The potential of carbon sequestration to mitigate against climate change in forests and agro ecosystems of Zimbabwe

Climate change adversely affects human livelihoods and the environment through alteration of temperatures, rainfall patterns, sea level rise and ecosystem productivity. Developing countries are more vulnerable to climate change because they directly depend on agriculture and natural ecosystem products for their livelihoods. Mitigation of climate change impacts includes practices that can store carbon (C) in soil and biomass thus, reducing concentrations of atmospheric carbon dioxide (CO2) and other greenhouse gasses. In addition, planted and natural forests that store large amounts of C, can become key resources for mitigating and reducing vulnerability to climate change, whilst infertile agricultural soils require large amounts of chemical and/or organic fertilisers to improve productivity. Increasing awareness about climate change mitigation has led to realisation of a need for sustainable land management practices and promoting soil C sequestration to reduce the greenhouse effects.

The C storage potential of agricultural soils is compounded by conventional tillage practices, covering large areas with only small portions of fields dedicated to conservation farming practices. Maintaining soil and crop productivity under these agricultural systems becomes a major challenge especially in rain-fed arid and semi-arid regions, characterised by long annual dry spells. Conservation tillage practices, such as no-till and reduced tillage, have been reported to increase soil organic carbon (SOC) stocks in agricultural systems as they reduce soil disturbance, whereas conventional tillage has been criticised for causing soil C losses, accelerating soil erosion and displacing of soil nutrients, despite benefits, such as reduced soil compaction, weed control and preparation of favourable seedbed, which have been reported under conventional tillage. The identification of appropriate agricultural management practices is critical for realisation of the benefits of Soil C sequestration and reducing emissions from agricultural activities.

This thesis was planned to improve our understanding on how tillage, fertilisation, tree planting or natural forest conservation can enhance C sequestration and thus mitigate climate change. The main goal was to quantify the influence of tillage, fertilisation and plantation forestry practices on C and N dynamics in bulk soil and density separated soil organic matter (SOM) fractions relative natural forest. Tillage treatments under reduced tillage (RP), no tillage (DS) and conventional tillage (CT) were compared with natural forests (NF) in sandy Haplic Arenosols and clayey Rhodic Ferralsols. Impacts of fertilisation were assessed from three fertility treatments; unfertilised control (control), nitrogen fertiliser (N Fert) and nitrogen fertiliser plus cattle manure (N Fert + manure) in conventionally tilled fields on Arenosols (sandy soil) and Luvisols (clayey soil) along two soil fertility gradients. Similarly, C and N storage in tree farming was studied using a Pinus patula chronosequence. Soil sampling followed randomised complete block design with four replications in agricultural systems and two replicates in each plantation age stands and natural forest. Sodium polytungstate (density 1.6 g cm-3) was used to isolate organic matter into free light fraction (fLF), occluded light fraction (oLF) and mineral associated heavy fraction (MaHF). Carbon an N were analysed by dry combustion and C and N stocks calculated using bulk density, depth and C and N concentration. The RothC model was used to match density separated fractions with conceptual model pools for agricultural and natural forest soils.

Findings from tillage studies showed significantly larger C and N stocks in natural forests than tillage systems despite the open access use of the natural forests. The C and N stocks were significantly lower in sandy than clayey soils. At 0–10 cm depth, SOC stocks increased under CT, RP and DS by 0.10, 0.24, 0.36 Mg ha−1yr−1 and 0.76, 0.54, 0.10 Mg ha−1yr−1 on sandy and clayey soils respectively over a four year period while N stocks decreased by 0.55, 0.40, 0.56 Mg ha−1yr−1 and 0.63, 0.65, 0.55 Mg ha−1yr−1 respectively. Under prevailing climatic and management conditions, improvement of residue retention could be a major factor that can distinguish the potential of different management practices for C sequestration.

Among the fertility treatments, there were significantly higher SOC and TON stocks under N Fert and N Fert + manure at 0-10 cm soil depth in Luvisols. Although this effect was not significant at 20-30 cm and 30-50 cm depth. On Arenosols, N Fert had highest C and N at all depths except at 0-10 cm. The storage of C and N on Luvisols, followed: control < N Fert < N Fert + manure whereas Arenosols had control < N Fert + manure < N Fert. Compared with control, N Fert and N Fert + manure enhanced fLF C on homefields and outfields by 19%, 24% and 9%, 22% on Luvisols and 17%, 26% and 26%, 26% respectively on Arenosols. Homefields on Luvisols, under N Fert and N Fert + manure had similar equilibrium levels, which were 2.5 times more than control.

Forests play a major role in regulating the rate of increase of global atmospheric CO2 storing C in soil and biomass although the C storage potential varies with forest type and plant species composition. In this research, storage of C and N were highest in moist forest and lowest in the Miombo woodland. In both natural and planted forests, above ground tree biomass was the major ecosystem C pool followed by forest floor’s humus (H) layer. The mineral soil had 45%, 31% and 24% of SOC stored at the 0–10, 10–30 and 30–60 cm soil depths respectively. Stand age affected C and N storage significantly having an initial decline after establishment recovering rapidly up to 10 years, after which it declined and increased again by 25 years. Average soil C among the Pinus compartments was 12 kg m-2, being highest at 10 years and lowest in the 1 year old stands. Organic N was also highest at 10 years and least at 25 years. The proportional mass of fLF and oLF in Miombo woodlands was similar while the other stands had higher fLF than oLF. The highest LF was in the moist forest. In the Pinus patula stands the fLF C contributed between 22−25%, the oLF C contributed 8−16% and MaHF C contributed between 60−70% to total SOC. Carbon in MaHF and oLF increased with depth while the fLF decreased with depth in all except the 1 and the 10 year old stands. Conversion of depleted Miombo woodlands to pine plantations can yield better C gains in the short and long run whilst moist forests provide both carbon and biodiversity. Where possible moist forests should be conserved and enrichment planting done in degraded areas to sustain them and if possible the forests can be considered as part of future projects on reduced emission from deforestation and degradation (REDD+). It is believed that REDD+ can promote both conservation and socio economic welfare, including poverty alleviation by bringing together the development of the forest and climate change link in African forests and woodlands. The focus on the monetary valuation and payment for environmental services can contribute to the attraction of political support for soil conservation. Developing countries therefore, need to formulate enabling economic and institutional land management policies that have positive impacts on poverty alleviation, food security and environmental sustainability.

Soil C models are used to predict impacts of land management on C storage. The RothC 2.63 model was used for estimating SOC stock under selected land management practices on the clayey and sandy soils of Zimbabwe. There is greater potential to store more C in clayey soils than sandy soils and in practices that receive more organic inputs. Results show that the RothC model pool of HUM + IOM is related to the measured MaHF from density fractionation and that the model can be used to estimate SOC stock changes on Zimbabwean agricultural and forest soils. The relationship between equilibrium levels estimated by the RothC model and those estimated using the Langmuir equation was good. A 1.5˚ C rise in temperature was found to cause the A and B systems on clayey soils to sequester more C. The results also show that, when holding all the other factors constant, the model is sufficiently sensitive to a rise in temperatures with sandy soils reaching an equilibrium much earlier than clayey soils. The modelling approach represents one of the most promising methods for the estimation of SOC stock changes and allowed us to evaluate the changes in SOC in the past period on the basis of measured data. However, since the data were obtained from short term experiments (4−9 years), further ground validation can be hampered by the lack of long-term experimental trials in the southern African region. The deficiency of adequate experimental sites also limits further work on model uncertainties. The understanding soil quality and dynamics however, helps to design sustainable agricultural systems, while achieving the urgently needed win-win situation in enhancing productivity and sequestering C.